EPIDEMIOLOGIC SYSTEMS IN CARDIOVASCULAR DECISION MAKING is the five-year research plan of Gerald T. O'Connor, Ph.D., an epidemiologist and medical information scientist. Central to medical decision analysis is the ability to predict clinical outcomes using epidemiologic data to obtain estimates of the effects of various disease processes and therapeutic or preventive interventions. In most medical decision making situations this requires the summarization of the results of several clinical research studies and an estimate of the utilities associated with each course of action. Significant research issues in both the epidemiologic methods used to combine these research studies and the methods used to assess utilities are the focus of this research plan. The overall goal of this research plan is to develop a high degree of expertise in the use of development and use of epidemiologic systems in medical decision making. This will be accomplished by the following specific aims: 1. Theoretical research in epidemiologic systems in medical decision making to focus on: - Epidemiologic aspects of medical decision making: using meta- analysis techniques to combine data from clinical studies. - Cognitive issues in medical informatics: physician and patient derived utilities in chronic disease management. 2. Applied research in medical decision making to focus on: - a formal decision analysis of the choice of mechanical heart valves or xenograft prostheses in the surgical management of valvular heart disease. 3. Educational initiatives in medical decision science: instructional activity in medical decision analysis.